EPIDEMIOLOGY AND HUMAN STREPTOCOCCAL DISEASE

Even before streptococci were known as human pathogens, diseases which we now know were caused by streptococci were controllable by accurately identifying the risk factor involved. But the challenge in those days was to convince others of the association. The epidemiological truth did not always prevail against widespread and entrenched ignorance. Two early examples, puerperal fever and epidemic sore throat, revealed the importance of identifying how disease is contracted and spread. In the case of puerperal fever it was to identify a particular obstetrician whom the disease followed like a shadow. In the case of epidemic sore throat it was to identify a dairy distributing contaminated milk.

Throughout the last century, before streptococci had been identified as pathogens, physicians in Europe and the United States suspected the contagious nature of puerperal fever, an acute, suppurative, and frequently fatal infection of women in child birth. Ignatz Semmelweis, a young obstetrical assistant in Vienna, suspecting the contagious nature of puerperal fever, conducted one of the earliest successful clinical trials to test his theory. He saved many lives in the process. In 1845, two obstetrical hospitals existed in Vienna, one for obstetrical teaching, the other where midwives were primarily responsible for deliveries. Death rates due to puerperal fever in the first clinic were 4 times the death rates in the second. Semmelweis proposed that the high mortality of the first clinic was due to students performing post mortem examinations on cases of puerperal fever, and then carrying toxic material directly to patients in the wards. By insisting that students wash their hands in "chloride of lime" following post mortem examinations, he dramatically reduced the death rate of the first clinic to the level of the second clinic. However, he was denounced by the medical authorities of the day, and received little recognition in his lifetime. The hand washing practice was discontinued by his superiors, and he was removed from his position. He himself died of the disease, from an infected cut obtained during post mortem examination of a puerperal fever victim (5).

In the United States, the Boston physician and novelist, Oliver Wendell Holmes, met similar opposition. Women frequently gave birth at home, attended by a visiting doctor or midwife. Holmes gathered medical records as far back as eighteenth century Britain to document, beyond reasonable doubt, a horrifying pattern of disease. Certain practitioners would suddenly be cursed with a series of frequently fatal cases of puerperal sickness in patients under their care. The doctor would rush from delivery to delivery, without washing his hands or changing his clothes, and in rapid succession his patients would contract the disease. In addition Holmes noted that a Scotsman, Dr. Gordon of Aberdeen, announced as early as 1795 that it was possible to predict which woman would become afflicted, after hearing which midwife attended the delivery. Holmes particularly noted the association of disease outbreaks following a physician performing an autopsy on cases of puerperal fever, or after treating patients with erysipelas (a septicemic streptococcal disease with severe skin lesions). He even records a physician who attended an autopsy of a case of puerperal fever, and then carried the pelvic viscera of the dead woman in his pocket to a class. He attended another woman in the evening without having changed his clothes. That woman was the first of a series he assisted who subsequently died. Because puerperal fever occurred sporadically in unattended deliveries, occurring on the average 3 times in every 1000 births, Holmes argued that a succession of 2 cases was highly unlikely due to chance, whereas the chance occurrence of a series of 3 cases was virtually impossible.

Many opposed his views bitterly, for obvious reasons. Countless physicians had encountered such series in their own practices, and winced to acknowledge that they were directly responsble for so many unnecessary deaths. Because of the wide scope of his documentation, probably assisted by his rhetorical and literary ability, he prevailed. His essay, The Contagiousness of Puerperal Fever, first published in 1843, is one of the earliest contributions to medical research by an American physician (4).

In December 1911, Chicago suffered a severe and widespread epidemic of sore throat and fever, afflicting over 10,000 people, for which no explanation could be made at the height of the epidemic. The severity of the outbreak was matched in only one other community in the state, the nearby town of Batavia, Illinois. An earlier outbreak in Boston had suggested a causal link between milk supply and epidemics of sore throat. Two physicians chose to investigate the possibility that the Chicago outbreak was a milk-borne disease (1). It quickly became clear that a disproportionate number of cases were associated with consumption of milk from one dairy (dairy X), which also supplied Batavia. Of all the cases of sore throat, 87% occurred in drinkers of milk from dairy X. Furthermore, the incidence of sore throat was 14 times higher in drinkers of milk from dairy X than from any where else. Cases of bovine mastitis had occurred on the farms supplying dairy X, simultaneously with cases of sore throats in milkers on those farms. But the most significant finding lay in the milk plant. Failures in achieving adequate pasteurization temperature were consistently recorded on days which immediately preceded the greatest numbers of new cases.

However, bacteriologic investigations of the outbreak ran into difficulties. A girl on one of the farms had suffered from sore throat, with arthritic complications, for many weeks but had not drunk any milk from the cows at the dairy. One cow on the same farm had mastitis. Similar organisms were cultured from the girl, the cow, and the throats of afflicted milk drinkers in Chicago. It was proposed that the same organism caused all these infections, and was named Streptococcus epidemicus (2). It was asserted, based largely on colony morphology that S epidemicus was distinct from the other streptococci isolated from the udder of cows, and distinct from other cases of septic streptococcal disease in humans, such as scarlet fever, erysipelas and streptococcal sore throat unassociated with milk. However, other epidemics of milk borne streptococcal disease were often associated with erysipelas and scarlet fever. Consequently the unique identity of S epidemicus was questioned (7). Attempts were made to classify other streptococci as distinct organisms linked to specific diseases, such as S scarlatinae from cases of scarlet fever (6). But distinguishing between the isolates could not be performed satisfactorily. There was no simple way to distinguish the many hemolytic streptococci which could be cultured from humans, nor clearly distinguish between human and animal strains. Investigators were severely hampered by an inability to confidently identify organisms, lacking any subdivisions by which streptococci from diverse sources could be grouped (3, 6).

REFERENCES

  1. Capps JA, Miller JL. The Chicago epidemic of streptococcus sore throat and its relation to the milk-supply. J Am Med Assoc 1912;58:1848-1852.

  2. Davis DJ. Bacteriologic study of streptococci in milk in relation to epidemic sore throat. J Am Med Assoc 1912;58:1852-1854.

  3. Dochez AR, Avery OT, Lancefield RC. Studies on the biology of Streptococcus I. Antigenic relationships between strains of Streptococcus haemolyticus. J Exp Med 1919;30:179-213.

  4. Holmes OW. The contagiousness of puerperal fever. (ed), The Works of Oliver Wendell Holmes IX: Medical Essays. Boston: Houghton, Mifflin and Co, 1892;103-172.

  5. Semmelweis I. Lecture on the genesis of puerperal fever. In:T. D. Brock (ed), Milestones in Microbiology. Washington: American Society for Microbiology, 1975;80-82.

  6. Sherman JM. The streptococci. Bacteriol Rev 1937;1:3-97.

  7. Williams AW, Gurley CR, Sobel E et al. Milk-borne septic sore throat and scarlet fever and their relation to beta hemolytic streptococci. J Bacteriol 1932;23:241-258.

Review question

How can you minimize outbreaks of the following diseases in the following populations?

  1. Parvovirus outbreaks in a veterinary hospital

  2. Feline Leukemia Virus spreading through the cat population

  3. Sendai virus in a herd (?) of laboratory mice

  4. Pasteurella in meat rabbits

  5. Influenza virus among college students

  6. Mammary tumors in the dog population?

  7. Tapeworms in the dog population

  8. Collie eye syndrome

  9. Hip dysplasia

  10. Dilatory cardiomyopathy in cats?